The basis for the design of the anti-HIV-1 drugs described in this project is the unique drug receptor characteristics of HIV-1 nucleic acids. HIV-1 has a highly folded single-stranded RNA genome, an RNA-DNA hybrid, and a DNA duplex that exist transiently in the host cell cytoplasm. In the first two years of this project, we have discovered the first simple organic molecules that bind strongly to RNA but very weakly to DNA, and molecules that bind selectivity to DNA-RNA hybrid duplexes. The anti-HIV-1 drugs in this proposal build on those discoveries. The HIV-1 RNA in critical gene control regions has duplex conformations with base bulges, loops and base-pair mismatches that are cis-acting elements in gene expression and are essential for HIV-1 replication. The folded regions of HIV-1 RNA differ from DNA in several characteristics: the RNA has a 2'-OH group in the minor groove that offers hydrogen bonding possibilities; the electrostatic potential of the RNA major groove is significantly more negative than either groove in DNA; and the steric characteristics of the RNA grooves are very different from those of DNA. Based on knowledge of RNA interactions, gained during the first two years of this proposal, we have designed molecules to take to take advantage of all of these features for selective interactions with RNA that will inhibit HIV-1 replication. A carboxyphenylphenanthridinium compound, for examples, binds insignificantly to DNA under physiological conditions but binds strongly to RNA. Considerable experimental evidence suggests that this molecule binds with high selectivity to duplex regions adjacent to base bulges such as those in HIV-1 RNA gene control regions (eg. TAR and RRE). Several unfused aromatic cations were found to bind quite strongly to both RNA and DNA, and new molecules are proposed that will also bind strongly to RNA but weakly to DNA. The design methods proceed stepwise to a new class of anti-HIV-1 drugs, RNA specific organic repressors (RASORS), that will inhibit HIV-1 gene expression and replication. The RASORS combine a duplex recognition unit, such as those described above, and a unit that binds to unpaired RNA bases such as those that exist at base bulges or loops. These units are joined by linkers that also enhance RNA binding. The duplex unit of RASORS will intercalate in TAR or RRE adjacent to base bulges or loops and place the unpaired base recognition unit of RASORS at an optimum position to interact with the unpaired bases. After the RNA interactions are optimized, additional groups will be added to the basic RASORS to selectively hydrolyze HIV-1 RNA. We have designed such a catalytic RASOR with a carboxyl that can hydrogen bond to an HIV RNA 2'-OH and transiently remove the hydrogen from the 2'-OH. The molecule has a cationic pyridyl function (for proton donation) adjacent to the phosphate at the same site so that RNA can be hydrolyzed in a ribonuclease type mechanism. This hydrolysis is selective for RNA and cannot occur with DNA. RASOR binding will direct the catalytic activity to the desired specific sites in HIV-1 RNA.